1. Introduction to single-cell RNA-seq
2. Quality control and normalization
3. Survey of downstream analysis methodology

• Goal: identify a set of genes with high variability attributable to biology (over and above technical variability)
• Useful for vizualization, dimension reduction, clustering, marker gene selection, etc
• Challenging without orthogonal measurement of technical variability (e.g. spikeins)
  • with spikeins: select genes with variance significantly above mean-variance trend in control genes
  • without spikeins: select genes with variance significantly above overall mean-variance trend in all endogenous genes (assumes variance of most genes is purely technical)

• Useful to summarize & visualize relationships between cells in low dimensional space
• Commonly used approaches:
  • PCA (Principal Components Analysis)
  • tSNE (t-distributed Stochastic Neighbor Embedding)
  • UMAP (Uniform Manifold Approximation and Projection)
• Clustering can be carried out on reduced dimensions, but with caution

• PCA attempts to extract the largest components of variation in the data
• Nonlinear methods such as tSNE and UMAP attempt to map points to a global coordinate system that preserves local structure
  • density & distance of points not preserved
  • better at visualizing rare subtypes

image source: http://carbonandsilicon.net/rblogging/2018/02/27/UMAP_plots

• Hierarchical: build and cut dendrogram based on pairwise distance matrix
• K-means: iteratively assign cells to nearest cluster center
• Density-based: clusters are defined as areas of higher density (e.g. DBSCAN)
• Graph-based: assumes data can be represented as a graph structure; doesn’t require estimating pairwise distance matrix (e.g. SNN-Cliq)

• Goal: automatically identify subpopulations of cell types/states
• Many methods adapted/developed for single-cell to account for:
  • high dropout rate
  • batch effects
  • high dimensionality

Andrews & Hemburg 2018 (https://doi.org/10.1016/j.mam.2017.07.002)

• Many clustering algorithms require specification of the number of clusters (e.g. K in K-means)
• Different algorithms may provide vastly different clusterings

Risso et al 2018 (https://doi.org/10.1371/journal.pcbi.1006378)

• Single clustering:
  • Average silhouette: how similar a cell is to its own cluster (average distance within) compared to other clusters (average distance between); not efficient for large numbers of cells \[s(b) = \frac{\bar{d}_{\text{within}}(b) - \bar{d}_{\text{between}}(b)}{max(\bar{d}_{\text{within}}(b), \bar{d}_{\text{between}}(b))}\]
  • Modularity score: difference between number of within-cluster edges to the expected number under the null (random edges)
• Comparing two clusterings
  • Adjusted rand index: similarity measure between two clusterings based on number of cells grouped accordingly and adjusted for random groupings

• Duo et al. 2018: https://doi.org/10.12688/f1000research.15666.2
• Bioconductor package to explore results: DuoClustering2018

• Traditional differential expression for bulk RNA-seq aims at detecting shifts in mean (fold change)
• Read counts \(y_{bg}\) typically modeled with a two parameter distribution (e.g. mean \(\mu_{bg}\), dispersion \(\alpha_g\))
• Recall DESeq2 model given size factors \(s_b\): \[ y_{bg} \sim NB(\mu_{bg}, \alpha_g)\] \[\mu_{bg}=s_b q_{bg}\] \[ log_2(q_{bg}) = x_{b.}\beta_g \]

Korthauer et al 2016 (https://doi.org/10.1186/s13059-016-1077-y)

• scater: visualization, quality control (Bioconductor)
• scran: normalization, doublet detection, batch effect correction (Bioconductor)
• SCnorm: normalization (Bioconductor)
• sctransform: normalization (CRAN)
• DropletUtils: removal of empty droplets (Bioconductor)
• Seurat: normalization (CRAN)

Zappia, Phipson & Oshlack 2018 (https://doi.org/10.1371/journal.pcbi.1006245)